The Economics of Hydrogen Production - ACS Symposium Series

1 The Economics of Hydrogen Production

Downloaded by 185.89.100.34 on October 18, 2016 | http://pubs.acs.org Publication Date: March 26, 1980 | doi: 10.1021/bk-1980-0116.ch001

D. P. GREGORY, C. L. TSAROS, J. L. ARORA, and P. NEVREKAR Institute of Gas Technology, 3424 South State Street, Chicago, IL 60616

Hydrogen is used today as a major chemical intermediate, as a reducing gas, and for a wide variety of special applications.

As fuel processors begin to deal with "heavier11 feedstocks, including coal and shale, hydrogen will play an increasing role in their conversion to clean and useful light hydrocarbon fuels. Consideration is currently being given to the use of hydrogen as a fuel for such applications as automobiles and aircraft, where its cleanliness and light weight offer advantages, and ultimately, as a possible replacement for natural gas and petroleum in the trans-

portation, storage, and delivery of energy derived from non-fossil (nuclear and solar) sources. (In the latter case, hydrogen is considered as a carrier of energy, analogous to electricity, rather than an energy source.)

The cost of producing hydrogen is aspects of its present and future use. turers must consider the changing cost stocks, the rising manpower costs, and capital.

a dominant factor in all Present hydrogen manufacand availability of feedthe increasing costs of

Decisions about the future uses of hydrogen require

accurate projections of the cost of manufacturing it by both conventional and still-developing methods.

There are four basically different methods of making hydrogen which must be considered: In the first category, a relatively reactive light hydrocarbon, such as naphtha or natural gas, is

reacted with steam, producing hydrogen derived partly from the hydrocarbon itself and partly from the steam. In the second category, heavy, relatively unreactive feedstocks such as coal and residual oil, are partially oxidized and reacted with steam,

producing hydrogen. In the third category, water alone is split electrochemically, producing hydrogen and oxygen, the needed electric power coming from one or more of a wide variety of raw energy sources. In the fourth category, water molecules are split in a complex chemical cycle which uses primarily heat as the energy input — all other materials of the cycle being continuously 0-8412-0522-l/80/47-116-003$06.00 © 1980 American Chemical Society Smith and Santangelo; Hydrogen: Production and Marketing ACS Symposium Series; American Chemical Society: Washington, DC, 1980.

4

hydrogen: production and marketing

recycled.

The heat for such a cycle would be produced from

nuclear or solar energy.

It is true to generalize that the cost of making hydrogen today is lowest for the first category and highest for the fourth; however, this does not mean that only the first category is in use today, although it appears to be the cheapest, nor does it mean

that the sequence of economic priorities will always be in today T s order.

In this paper, we have not attempted to answer the question

of "How much does hydrogen cost?" or of "Which process is the most


economic?" Rather, we have set out, on a parametric basis, a range of hydrogen costs for a selected variety of production

processes covering all of the categories mentioned. We have shown the effect of some major variables, such as feedstock price and by-product credits, where applicable. This allows for an understanding of the criteria underlying any more-detailed analyses of hydrogen production costs. Additionally, in the scope of this paper, we have not attempted a detailed process description and economic analysis of each process considered. We have taken economic data available in the published literature and converted the data to a common basis in line with the guidelines (shown in Table I ) . These are largely those supplied by the organizers of this symposium* with the following additions or modifications: •

Interest during construction added:

• •

investment Land: not included Fuel: same as feedstock

15% of facilities

•

Return on Investment, or cost of capital:

20% before

taxes = 10% after tax at 50% tax rate.

Table II gives a summary comparison of the major economic elements

for each process. Base plant capacity is 100 million SCF/day of

hydrogen; investment is representative of estimated mid-1979 costs. Process Descriptions

Hydrogen By Steam Reforming. Historically, steam reforming is the most widely used and the most economical process for producing hydrogen from light hydrocarbon gases or from naphthas. The process is based upon the well-known reaction of steam and hydrocarbons, which is carried out over a nickel catalyst. With methane, the reaction is CH4 + H20 = CO + 3H2; the reaction is highly endothermic, and is carried out in a reformer tube furnace

fueled by the feedstock. Exit temperatures range from 1300° to

1600°F. The nickel catalyst is poisoned by sulfur; thus, sulfur is removed from the feed. Figure 1 is a block flow diagram showing the basic process steps with natural gas as the feed. Reformer effluent is cooled by waste heat recovery, which generates reaction steam plus some by-product steam. The carbon • ACS Spring Meeting:

"Production and Marketing of Hydrogen,

Current and Future," April 5-6, 1979, Honolulu, Hawaii.

Smith and Santangelo; Hydrogen: Production and Marketing ACS Symposium Series; American Chemical Society: Washington, DC, 1980.

1. Gregory et al.

Economics of Hydrogen Production

Table I. Project Life:

ECONOMIC EVALUATION GUIDELINES

20 years

Operating Factor:

330 days/year

Capital Investment (mid-1979 basis) Facilities Investment (F.I.) = As obtained from literature and adjusted to mid-1979 basis Interest During Construction = 0.15 F.I. Startup Expense = 0.02 F.I. Working Capital = 60 days feedstock utilities and cash supply Total Capital Investment = Sum of the Above Downloaded by 185.89.100.34 on October 18, 2016 | http://pubs.acs.org Publication Date: March 26, 1980 | doi: 10.1021/bk-1980-0116.ch001

Annual Operating Cost Feedstock

Natural Gas Wyoming Coal Montana Coal

$2/106 Btu $20/ton ($1.17/106 Btu) $17.7/ton ($1.00/106 Btu)

Residual Oil

$15/bbl

Utilities

Raw Water ($/103 gal) 0.50 Power ($/kWhr) Cooling Water

0.030 0.05

($/103 gal) Fuel ($/106 Btu) same price as for feedstock

Cash Supply

Operating Labor

$18,000/man-year

Operating Labor Supervision

15% of operating labor

Maintenance

Labor

2% of facilities investment

Supervision

15% of maintenance labor

Materials


Administrative and Support 20% of all other labor Labor

Payroll Extras/Fringe

20% of all other labor

Benefits, etc.

Insurance


General Administrative Expenses 2% of facilities investment

Taxes (Local, State, and

50% of net profit

Federal) (No investment Tax Credit)

Depreciation

Straight line, 5%/year

Return on Investment

20% before taxes (0.2 X C)

(cost of capital) Gross Revenue Required By-Product Credits

Sum of the Above

Sulfur Steam

$10/long ton $3/1000 pound

Power Oxygen

2ç/kWhr $10/ ton

Net Revenue Required

= Gross Revenue Required — By-Product

Hydrogen Price, $/103 CF

= Net Annual Revenue Required

Credits

Annual Hydrogen Production


5

6

HYDROGEN: PRODUCTION AND MARKETING

w

>, ow >-i P-i 4-> CO o

3FeO + H20

Oxidizer:

3FeO + H20 -> Fe304 + H2

The oxidizer effluent contains 37% hydrogen and 61% steam

plus small amounts of nitrogen and carbon oxides. Condensation of the steam leaves a gas containing 95.9% hydrogen, 1.6% carbon oxides, and 2.5% nitrogen. No carbon monoxide shift or acid-gas scrubbing are needed.

A cleanup methanation step reduces carbon

oxides to 0.2% followed by drying and compressing to 325 psig to give product gas. The carbon monoxide and hydrogen in the producer gas are not

completely converted in reducing iron oxide. Heating value plus sensible heat at 1520°F in the gas exiting from the reductor represent 54% of the input coal fuel value. Part of this energy, 15% of coal fuel value, is utilized in the plant to compress air and generate steam. A larger amount is used to generate electric power. After dust removal with cyclone separators and electrostatic precipitators, the effluent gas, at 1520°F and 365 psia, is used to fuel a gas turbine system.

The gas is burned in a

combustor with excess air at 290 psia. (A 75 psi control valve loss is assumed.) The effluent from the combustor, at 2400°F, is expanded to 16.5 psia and 1198°F in the gas turbine. A portion


12


AC

i î 00 < Q.

A z

!1

r

es

i


T v ' ¦¦ TT

v ' /"VT ' uj^VJ / K \.[ ( j

I

'

s—^'

/'y ( /

T^T

T

,-?- 32

"£• i l

1

?

ühS

|z

1 1 1 T^lüirii:

S I i

S

CO


8»

s •§,

I»

1. Gregory et al.


13

of the expansion power is used to drive the combustor and producer air compressors. The remaining shaft power is converted to electric power in a generator. Currently, the maximum temperature

range is 1800° to 2000°F, but gas turbines with an inlet temperature of 2400 °F are expected to be available by the time this process becomes commercial.

The expanded gas is used in a steam-power cycle to generate electric power from 2400 psig, 1000°F steam generated in a wasteheat boiler, and to generate a portion of the required process steam for the oxidizer. The cooled combustor gas leaves the steam


cycle waste-heat boiler at 350°F.

A total of 1325.4 MW of power

is generated in addition to the shaft power used for air compression; approximately 7% to 8% of this energy is used in the plant for motor drives, etc. With this design adapted to an output of 100 million SCF/day of hydrogen, the by-product electric power

amounts to 158 MW.

The base credit value assumed for this is

2ç/kWhr. Basic investment and operating cost information were scaled

down to 100 million SCF/day from a larger IGT design (2) . The total capital cost, according to the guidelines set in Table 1, is $237 million. At the base coal cost of $17.7/ton ($1.00/106

Btu), hydrogen price is $2.56/103 CF, or $7.84/106 Btu. Hydrogen By Water Electrolysis In water electrolysis, alternating current is rectified to

direct current and fed to a battery of electrolyzer cells. Hydrogen and oxygen are released separately and the gases are manifolded from the cell batteries into separators, where water vapor and excess electrolyte are removed. The system costs included feedwater treatment and hydrogen compression to 400 psi. A block flow diagram is shown in Figure 5. Two cases were examined for the production of water electrol-

ysis.

Data were taken from Reference 0_) and adjusted to mid-1979

levels in accordance with Table 1.

The costs of "current tech-

nology" electrolysis were averaged in Reference (JL) from information provided by Lurgi, Electrolyser Corp., General Electric, and

Teledyne Isotopes. An advanced electrolyzer design, based upon the General Electric Solid Polymer Electrolyte (SPE) design, was also addressed as the second case.

In the source reference, Chem Systems have assumed an overall efficiency, power-to-hydrogen, of 75.7% and 77.6%, respectively,

for current and SPE technology.

The goals of the General Electric

SPE program are to achieve a 90% efficient system — already demonstrated on the bench scale — and to reflect this, we have taken a third case in which the hydrogen costs were recalculated using a 90% efficiency. Assuming that such an efficiency might

be achieved commercially, both capital and operating costs are lowered:

The facilities investment and electric power require-

ments were reduced by a factor of 77.6/90 = 0.862 from the original design.



14


_____________ ÇfL J

r-II

T--°

I I

ELECTROLYSIS MODULES

,

P0WER IN

O2/H2O /WWWWWvJ

»

ELECTRICAL

l| CONVERTER |l

,

f

I

»

Z

2 Q

H2/H2Q

LZ^ EXCHANGER VWVWWWW.J

MM AIR

[gl n

HEAT

i o2/h2o Lpri — WATER

CIRCULATING PUMP

8^ ~~

¦

SEPARATOR

—p

H20

'

tH2°

' '

REG. „M

'FEED

2

0UT

WATER PUMP

JT SEPARATOR H2/H20 IPiI REG. H2 I|

Q m DISTILLED ^"H2° ,N "2 OUT

Figure 5. SPE electrolytic hydrogen generating plant — simplified schematic


1. Gregory et al.


15

Thermochemical Hydrogen Production If hydrogen is to be used as a carrier of energy produced from nuclear or solar furnaces, there is an incentive to develop processes for the thermal dissociation of water. The problem of splitting water using thermal energy alone within the temperature constraints of practical nuclear and solar heat sources has been addressed by the concept of thermochemical water splitting. In

this concept, a multi-stage cyclic chemical process would be


operated in which water and heat are the only inputs and hydrogen, oxygen, and degraded heat are the only outputs. To date, this technology is receiving wide attention, but it is still in its infancy and is thus difficult to make sound economic projections.

Literally hundreds of cycles have been proposed and analyzed mainly on the grounds of efficiency and operability. Some of the more promising cycles involve both thermal and electrochemical steps and are known as "hybrid" cycles. In these, some of the input heat is used to produce electric power which in turn is used to drive an electrochemical reaction. For a few promising cycles, stepwise, bench-scale demonstrations of the chemistry

have been performed, and some flowsheet analysis has been carried out. For the purposes of this paper, one cycle has been selected on the basis of having had more economic study performed on it

than most of the others. The cycle is known variously as the "Westinghouse Sulfur Cycle" and the "EURATOM Mark II-V6 Cycle." It is a thermal-electrochemical hybrid cycle as follows:

2H20 + S02 -> H2S0i+ + H2 (electrolysis) H2S0I+ ¦* H20 + S03 (thermal) S03 -> S02 + ^02 (thermal) The process is assumed to operate on thermal energy delivered as high-temperature helium from a high- temperature, gas-cooled nuclear reactor (VHTR) operating at about 1742°F. An economic analysis was carried out by Knoche and Funk (U) based upon a flowsheet developed by Westinghouse. This analysis indicated a projected hydrogen cost of about $6.50/10° Btu based upon the following fundamental assumptions:

•

Cost of nuclear heat:

•

Process efficiency:

• •

$1.50/10^ Btu

45%

Capital cost of non-nuclear portion of the plant: Plant size: 360 X 106 SCF/day hydrogen

$425 X 106

• Nuclear reactor size: 3345 MW (thermal). Because of the considerable uncertainties in all the assumptions

made, this projected hydrogen cost should not be rigorously compared with the cost projections we have made for the other processes discussed in this paper. Moreover, we have not reworked Knoche and Funk's economics to conform with the guidelines of

Table 1. (The assumptions they made were an 80% stream factor, utility financing, 12^% capital recovery factor, and mid-1976 dollars.)


16


A completely independent assessment of the same cycle was made by Broggi et al. (5) Their estimated hydrogen cost was

about $9/106 Btu based upon process heat supplied at $2.50/106 Btu.

The selected cost of nuclear process heat appears to be somewhat arbitrary, although in Knoche and Funk's case, they derived the cost from a composite of fuel, operating, maintenance, and investment components.

If solar heat is considered for a thermochemical hydrogen production process, two additional variables must be considered. The temperature limitations on a solar furnace are controlled by factors other than those for a nuclear reactor, and it is possible Downloaded by 185.89.100.34 on October 18, 2016 | http://pubs.acs.org Publication Date: March 26, 1980 | doi: 10.1021/bk-1980-0116.ch001

to consider temperatures in excess of 1742°F, normally considered to be the limit of HTGR technology. However, it is important to know how the cost of heat varies as a function of the delivery temperature, and no data for this variation are available for solar furnaces in the 1292° to 2732°F range. Second, because a

solar furnace will only operate intermittently, the plant factor will be considerably less than 100%.

No analysis has been

published to date for the sensitivity of cost of a thermochemical hydrogen process to plant factors in the 20% to 40% range. Although the production of hydrogen from solar thermal heat is an interesting and potentially important technology, it has not yet

progressed to the stage where even educated guesses can be made as to its cost.

Comparison Of Processes Table III presents a detailed summary of investment and annual

operating costs for the different processes. These are "base cases" from which the sensitivity analysis of various factors has been derived. These can be considered as representative of current costs for the purpose of process comparison. Sensitivity Analysis. The sensitivity of hydrogen price to

the following variables was calculated: raw materials cost, plant capital cost, stream factor, plant size, and by-product credits. The base case assumptions are shovn in Table III. The methodology used for these sensitivity calculations was as follows: •

Raw Materials Cost:

We assumed two or more raw material costs

and then calculated the hydrogen price using the following procedure for each raw material cost:

Raw Material

Residual Fuel Oil

Wyoming Coal

Process

Partial Oxidation

Koppers-Totzek

____________ Raw Material Price

Below

Base

Above

$10/bbl

$15/bbl($12. 35/10b Btu)

lc/kWhr — —

3c/kWhr ($8. 79/106 Btu) 5ç/kWhr $17. 7/ton($l .00/106 Btu) $1.50/106 Btu $2.00/106 Btu $4.00/106 Btu

$10/ton

$20/ton($l . 17/106 Btu)

$20/bbl

$30/ton

Process

Electricity Natural Gas

Montana Coal

Electrolysis Steam Iron Steam Reforming


1. Gregory et al.

Economies of Hydrogen Production

17

OO o o c cn • • • • • Os •> ~* ^h *-h m os os o r-^-HCNOOCN

m os

o r-~ oo cn

so ro so ro o 00 CN 00 OO so

-

'°

^

^v ,_, ca £

jo

,

°

2 «rt 24L ^ ^ °* ^ r° 5™^ O O • • • • • ro TO 2 o — i so os co r^ r- rw-ho '\omn coro S3 ¦ 3HW--H |cn cn

M

^ on » s 4-1

^

4J

co

i---

«m

i|

x

|

»

i

3 CN Os Os dlro

Os

. . . . . m so — < cn r^ oo oo

The Economics of Hydrogen Production - ACS Symposium Series

Recommend Documents